摘要
灰度图像分割问题一般采用传统的最大类间方差法来解决,但是类间方差方法计算量大,不适合实时图像处理。为了解决上述问题,提出了一种改进型遗传退火的阈值分割算法。算法的整个运行过程由冷却温度进度表控制,使用经典的最大类间方差法作为遗传算法的适应度函数,再根据M etropolis准则判断产生的新解是否被接受,从而求得灰度图像的一个最佳阈值。图像分割的仿真结果表明,与传统的最大类间方差法相比,计算量不大,算法具有很强的全局优化搜索能力,由于算法效率高,收敛速度快,适用于实时性的灰度图像处理。
In general the traditional Otsu method is used to solve the gray image division problem, but this meth- od is not suitable for real - time image processing because of large computation. In order to solve this problem, this paper mainly proposes a threshold selection algorithm based on the GASA method. The whole running process of this algorithm was controlled by the temperature cooling schedule, with the classical Otsu method being used as the fitness function of the genetic algorithm. The Metropolis principle is used to determine whether the new solution is available. After several rounds of computing, an optimal threshold value was obtained. The image simulation result indicates that this algorithm has stronger optimal searching ability and provides higher efficiency and faster convergence speed, which makes it an appropriate solution to implement the real -time gray image processing.
出处
《计算机仿真》
CSCD
北大核心
2010年第4期210-214,共5页
Computer Simulation
基金
上海市科委自然科学基金(08ZR1415300)
上海市科委研发平台(08DZ2290900)